Welcome to The 1st Workshop on Efficiency, Security, and Generalization of Multimedia Foundation Models at ACM Multimedia 2024!
The rapid progress in foundation models has enhanced the capabilities of multimedia models across a broad spectrum of tasks. Despite their exceptional performance, deploying these models in practical settings raises several concerns, particularly regarding efficiency, security, and generalization. As the utility of foundation models in multimedia topics becomes increasingly evident, addressing these issues is crucial. This workshop focuses on these critical aspects in foundation models, where the scope of the foundation model encompasses a wide range of domains such as vision, language, speech etc., with an emphasis on multimedia tasks and multi-modality methods.
Therefore, we solicit original research papers in (but not limited to) the following topics:
Efficiency
- Efficient network design in foundation models
- Training efficiency of foundation models
- Inference efficiency of foundation models
Security
- Adversarial robustness of foundation models
- Privacy and memorization in foundation models
- Trustworthiness and alignment of foundation models
Generalization
- Generalization across tasks
- Generalization across data
- Generalization across modalities
Important Dates
| Submission Open | May 11, 2024 (AoE) |
| Submission Deadline | |
| Decision Notification | August 5, 2024 (AoE) |
| Camera-Ready Deadline | August 19, 2024 (AoE) |
| Workshop Date | Oct 28 PM, 2024 |
Schedule
The following schedule is tentative and will be confirmed closer to the workshop:
| Time | Topic | Speakers/Details |
|---|---|---|
| 13:30 | Start | |
| 13:30-13:40 | Welcome Opening | |
| 13:40-14:10 | Invited Talk | Dr Piotr Koniusz (Data61, CSIRO) |
| 14:10-14:40 | Invited Talk | Prof Jiebo Luo (University of Rochester) |
| 14:40-15:10 | Invited Talk | Prof Phoebe Chen (La Trobe University) |
| 15:10-15:30 | Coffee Break | |
| 15:30-16:00 | Industry Talk | Dr Luoqi Liu (MT Lab, Meitu) |
| 16:00-16:15 | Paper Presentation | Object-Driven Human Motion Generation from Images |
| 16:15-16:30 | Paper Presentation | Rethinking the Role-Play Prompting in Mathematical Reasoning Tasks |
| 16:30-16:45 | Paper Presentation | ITCD: Image to Text Translation for Classification by Diffusion Models |
| 16:45-16:50 | Closing and Awards | |
| 16:50 | End |
Invited Speakers
Piotr Koniusz
Data61, CSIRO
Jiebo Luo
University of Rochester
Phoebe Chen
La Trobe University
Luoqi Liu
MT Lab, Meitu
Organizers
Daochang Liu
The University of Sydney
Minjing Dong
City University of Hong Kong
Yasmeen George
Monash University
Chang Xu
The University of Sydney
Sponsors
Contact
Contact the organizers at mm2024-esgmfm@googlegroups.com
Call for Papers
Submission Deadline: July 19, 2024 Extended to July 29
Submit Platform: OpenReview
We welcome submissions of research papers, demos, datasets, and position papers within the workshop’s scopes. The submission guideline follows the main conference site of ACM Multimedia 2024, including the formatting guideline and submission policies. The review process for this workshop will be “double-blinded”. Submissions should be of up to 4-page length in ACM-MM format, plus up to 1 additional page for the references.
All papers will be peer-reviewed by at least three experts in the field, regarding the relevance to the workshop, scientific novelty, and technical quality. Accepted submissions will be presented via oral or poster sessions. All accepted papers will be published in the ACM Multimedia proceedings in the ACM Digital Library. We will select from the accepted papers the Best Paper Award, which will be announced during the workshop.
Submit your manuscripts through OpenReview. All the authors need to create a profile on OpenReview. New profiles created without an institutional email will go through a moderation process that can take up to two weeks. New profiles created with an institutional email will be activated automatically.